Abstract
Background and Purpose Current auto-segmentation models of brain structures, UNets and nnUNets, have limitations, including the inability to segment images that are not represented during training and lack of computational efficiency. 3D capsule networks (CapsNets) have the potential to address these limitations.
Methods We used 3430 brain MRIs, acquired in a multi-institutional study, to train and validate our models. We compared our CapsNet with standard alternatives, UNets and nnUNets, based on segmentation efficacy (Dice scores), segmentation performance when the image is not well-represented in the training data, performance when the training data are limited, and computational efficiency including required memory and computational speed.
Results The CapsNet segmented the third ventricle, thalamus, and hippocampus with Dice scores of 95%, 94%, and 92%, respectively, which were within 1% of the Dice scores of UNets and nnUNets. The CapsNet significantly outperformed UNets in segmenting images that are not well-represented in the training data, with Dice scores 30% higher. The computational memory required for the CapsNet is less than a tenth of the memory required for UNets or nnUNets. The CapsNet is also more than 25% faster to train compared with UNet and nnUNet.
Conclusion We developed and validated a CapsNet that is effective in segmenting brain images, can segment images that are not well-represented in the training data, and are computationally efficient compared with alternatives.
Competing Interest Statement
Acknowledgements: Arman Avesta is a PhD Student in the Investigative Medicine Program at Yale which is supported by CTSA Grant Number UL1 TR001863 from the National Center for Advancing Translational Science, a component of the National Institutes of Health (NIH). The contents of this article are solely the responsibility of the authors and do not necessarily represent the official view of NIH. The data used in this article were obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). The investigators within the ADNI contributed to the design and implementation of ADNI but did not participate in the analysis or writing of this article. Disclosures: Arman Avesta: Arman Avesta is on the trainee editorial board of Radiology: Artificial Intelligence. Journal policy recused the author from having any role in the peer review of this manuscript. Research Funding: CTSA UL1 TR001863 from the National Center for Advancing Translational Science. The MedNet, Inc, American Cancer Society, National Science Foundation, Agency for Healthcare Research and Quality, National Cancer Institute, ASCO Potential Conflict of Interest: Arman Avesta holds securities at Hyperfine Inc. Yongfeng Hui: The publication was written prior to Yongfeng Hui joining Amazon. Mariam Aboian: Research funding: KL2 TR001862 from the National Center for Advancing Translational Science and NIH roadmap for Medical Research. The contents of this article are solely the responsibility of the authors and do not necessarily represent the official view of NIH. James S. Duncan Patents, Royalties, Other Intellectual Property: Systems, Methods and Apparatuses for Generating Regions of Interest from Voxel Mode Based Thresholds, Publication No: US20190347788A1, application No. 15/978,904. Filed on May 14, 2018, Publication Date: November 14, 2019. Inventors: Van Breugel J, Abajian A, Treli- hard J, Smolka S, Chapiro J, Duncan JS and Lin M. Joint application from Philips, N.V. and Yale University. US Patent 10,832,403 (2020) Harlan M. Krumholz Employment: Hugo Health (I), FPrime Stock and Other Ownership Interests: Element Science, Refactor Health, Hugo Health Consulting or Advisory Role: UnitedHealthcare, Aetna Research Funding: Johnson and Johnson Expert Testimony: Siegfried and Jensen Law Firm, Arnold and Porter Law Firm, Martin/Baughman Law Firm Sanjay Aneja: Sanjay Aneja is an Associate Editor for JCO Clinical Cancer Informatics. Journal policy recused the author from having any role in the peer review of this manuscript. Consulting or Advisory Role: Prophet Consulting (I) Research Funding: The MedNet, Inc, American Cancer Society, National Science Foundation, Agency for Healthcare Research and Quality, National Cancer Institute, ASCO Patents, Royalties, Other Intellectual Property: Provisional patent of deep learning optimization algorithm Travel, Accommodations, Expenses: Prophet Consulting (I), Hope Foundation Other Relationship: NRG Oncology Digital Health Working Group, SWOG Digital Engagement Committee, ASCO mCODE Technical Review Group
Funding Statement
Arman Avesta is a PhD Student in the Investigative Medicine Program at Yale which is supported by CTSA Grant Number UL1 TR001863 from the National Center for Advancing Translational Science a component of the National Institutes of Health (NIH). The contents of this article are solely the responsibility of the authors and do not necessarily represent the official view of NIH. This work was supported by the Radiological Society of North America Fellow Research Grant Number RF2212RO.
Author Declarations
I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.
Yes
The details of the IRB/oversight body that provided approval or exemption for the research described are given below:
The Image and Data Archive (IDA): ida.loni.usc.edu We downloaded 3430 T1-weighted brain MRI volumes belonging to 841 patients from the Image and Data Archive (IDA). These patients were enrolled in the multi-institutional Alzheimer Disease Neuroimaging Initiative (ADNI) study: adni.loni.usc.edu
I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.
Yes
I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).
Yes
I have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable.
Yes
Footnotes
nnUNet experiments were added.
Data Availability
All data produced are available online at: adni.loni.usc.edu
Abbreviations
- ADNI
- Alzheimer’s disease neuroimaging initiative
- CapsNet
- capsule network
- CPU
- central processing unit
- CT
- computed tomography
- GB
- giga-byte
- GPU
- graphics processing unit
- MRI
- magnetic resonance imaging